Skip to content
View commentaries on this research

This is a plain English summary of an original research article

One in ten inpatients in Welsh hospitals experienced an adverse event of some sort, about half of which were potentially preventable. This is very similar to the rate in other countries. The figures come from a review of 4,388 patient records between October 2010 and March 2013.

Welsh hospitals were using a screening tool at the time called the Global Trigger Tool to estimate rates of adverse effects. The tool surveys case notes of 20 inpatient records each month. Alongside the trigger tool process, researchers reviewed case notes using a more comprehensive two-step process – nurse screening followed by physician review – to compare how different the estimates were and to better understand how many events were preventable and the degree of harm. The two-step process found 10.3% of patients had an adverse event, while the tool found 8% adverse events.

Hospitals which carried out more surgery had higher rates of adverse events. Use of this trigger tool has since been dropped in Wales and a new “hybrid” system of adverse event identification is being developed.

Why was this study needed?

Global estimates of rates of adverse events range from about 8% to 12% of inpatient admissions – meaning that about 1 in 10 patients will have something happen to them in hospital that will harm them. However, measuring rates of adverse events is not easy. Methods mainly rely on ‘post-hoc’ analysis of patient notes after an adverse incident, although not all incidents are recognised or recorded at the time.

Up until recently, Welsh hospitals used the tool to measure adverse events. Twenty randomly selected inpatient case notes records per month per organisation are reviewed to estimate rates of adverse effects. The aim of this study was to compare the rate estimated by the tool to the rate estimated by a more comprehensive, two-stage review of case notes performed over four years.

What did this study do?

The rate of adverse events was estimated in two hospitals from each of the six Welsh health boards using two methods. One method used the Global Trigger Tool and the second used a comprehensive, two-stage review of case notes.

The Trigger Tool system: Hospitals reviewed case notes from 20 randomly-selected adult patients each month that had spent more than 24 hours in hospital. This involved a short, 20 minute review of notes using a checklist of criteria to identify actions that could have or did cause patient harm.

The two-stage review: Researchers subjected case notes, covering 4,388 inpatient episodes, to a two-stage check. Nurses did an initial checklist screen, passing notes with potential adverse events to a physician for a second review. The physicians also looked at whether adverse events could have been prevented.

What did it find?

  • The two-stage process found more adverse events than the trigger tool system – 10.3% (95% confidence interval (CI) 9.4% to 11.2%) versus 9% (95% CI 8.82% to 9.18%)..
  • The percentage varied between hospitals, with results ranging from 7.9% to 16.1%. Individual hospitals were not identified in the report.
  • Factors linked to higher levels of adverse events included longer hospital stay and being admitted for planned surgery. Hospitals which carried out more surgery had higher rates of adverse events.
  • Just over half (51.5%) of all adverse events were thought to be preventable (95% CI 46.88% to 56.12%) using the two-stage process.
  • An extension of this study looked at problems in care linked to the preventable adverse events. These included failure in clinical monitoring (33.5%), failure in infection control (29.4%), problems from surgical or invasive procedures (21.2%) and problems from administering drugs or fluids (18.7%).

What does current guidance say on this issue?

The Global Trigger Tool is no longer used in Wales. NHS England has a Serious Incidence Framework for the reporting and management of adverse events, but this does not include routine surveillance for and monitoring of adverse events.

What are the implications?

The study provides reassurance that rates of adverse events are in line with those reported in other parts of the UK and internationally. Any potentially preventable adverse event is of concern and use of real-world data (data collected as part of routine care) is becoming easier as hospitals use electronic patient records more, potentially this could lead to earlier identification of risks and prevention of harm.

The report identifies areas of care which most often led to an adverse event. This may provide useful lessons for hospitals to focus on.

The resources for implementing any safety system also need considering. While the tool is no longer used in Wales, the two-stage process was resource-intensive. The team who carried out this research is developing a replacement system for ongoing adverse event monitoring in Wales using the concept of risk signatures and using incident reports.


Citation and Funding

Mayor S, Baines E, Vincent C, et al. Measuring harm and informing quality improvement in the Welsh NHS: the longitudinal Welsh national adverse events study. Health Serv Deliv Res. 2017;5(9).

This project was funded by the National Institute for Health Research Health Services and Delivery Research programme.



NHS England. Serious Incident Framework. London: NHS England; 2015.

Produced by the University of Southampton and Bazian on behalf of NIHR through the NIHR Dissemination Centre


  • Share via:
  • Print article


These characteristics and definitions varied depending on the tools that were used.Adverse event:  An injury to a patient related to medical management, in contrast to complications of disease. Medical management includes all aspects of care, including diagnosis and treatment, failure to diagnose or treat, and the systems and equipment used to deliver care. Adverse events may be preventable or non-preventable.Harm: (two stage process) An unintended injury or complication that results in disability at discharge, death or prolonged hospital stay and is caused by health-care management rather than the patient’s underlying disease. Includes errors of commission and omissionHarm: (global trigger tool) Unintended injury resulting from or contributed to by medical care that requires additional monitoring, treatment or hospitalisation, or that results in death. Includes errors of commission only. 
Back to top